Cap table as editorial asset — Peter Walker on Carta's insights function
How a B2B SaaS company turned privileged data into a marketing moat — and what the 75% exercise-rate decline actually says about the cycle.
Peter Walker spent 2020 running data visualisation for the COVID Tracking Project out of *The Atlantic* — charts he built ended up everywhere on the internet for a year. The experience reframed what data could do for a B2B company: not just power internal dashboards, but become the marketing motion itself. He joined Carta in 2021 as Head of Insights, sitting under the CMO, and built the function from cleaning data for the first three to six months to running the most-cited primary-data practice in the early-stage ecosystem. Carta sits on roughly 43,000 startup cap tables and 2,500 venture funds; Peter's job is to turn that position into shareable signal.
The spine of the conversation is the cap table as editorial asset. Carta’s structural advantage isn’t dashboards inside the product — it’s the willingness to publish aggregated, anonymised, contextual analysis that early-stage founders and investors actually use. The 75% no-exercise rate on departing-employee equity is the sharpest worked example: a number that would have been hidden in a dataset is, in Peter’s framing, a piece of insight that makes the cycle legible. Insight is the answer to a question worth asking. Data storytelling is how you make it land.
Insight first, storytelling second — and why most companies do neither
The two-word vocabulary matters. An insight is the aggregated result of a dataset that tells you to act or not act — median Series A valuation, useful at the moment a founder is negotiating that round. Data storytelling is the craft of turning the spreadsheet into charts, headlines, and narratives that make the insight legible to a specific audience.
A lot of marketers do themselves a disservice by not understanding data as much as they should.
Peter’s complaint cuts both ways. Marketers under-consume data, treating it as ROI scaffolding rather than as the substrate of the work. Data teams over-consume their own data, building things technically correct and narratively dead. The combination — someone fluent enough in both to ask a question worth asking and present the answer so a non-analyst feels the shape of it in sixty seconds — is rare enough that it becomes a marketing moat. The headline-as-takeaway trick Peter teaches is the simplest example: don’t write number of new deals per quarter above the chart; write Q4 deals declined because X so the conclusion survives the scroll.
The funding winter, asymmetric by stage
The state-of-the-market section is the part most founders would clip first.
The funding winter is thawing, but is still rather chilly.
The numbers anchor everything. Carta startups raised about $220 billion in 2021, the all-time peak; the same population raised about $66 billion in 2023, a roughly two-thirds collapse. The Q1 2024 thaw was real but modest — incremental over Q4, not a regime change. The asymmetry is what matters: seed funding is down about 30–35% from peak, Series D is down about 80%. The late-stage drought is the worrying signal because dollar volumes are large and the IPO window has been shut since early 2023 — Klaviyo and Reddit are the standard-bearers, and that’s a thin shelf. Peter’s prediction at recording: company shutdowns keep rising through H1 2024 as 2023’s bridge rounds stop and VCs concentrate on new investments. January 2024 was the worst month for shutdowns in Carta’s dataset.
Exercise rates — the bearish signal hiding in plain sight
The single sharpest insight in the conversation is the one most founders haven’t internalised. About 75% of departing employees across Carta let their incentive stock options lapse during the 90-day post-departure exercise window. The only compensation components that became liquid for them were salary and bonus.
Things are a lot harder than they appear on the surface.
The mechanic compounds. Down valuations mean exercising is a cash outlay on a security potentially worth less than the strike price. Companies staying private 12 to 14 years (versus 8 to 10 historically) lock that cash up for longer. The option pool returns to the company and gets reissued — often in smaller packages to the next hires. The full compensation picture, from the same dataset: salaries are roughly flat since November 2022 (about 0.5% higher, well behind inflation); equity is down about 36% in share count for a January-2024 hire versus the same role in November 2022. The composition has shifted toward cash. That feels good in a downturn and worse if the company eventually succeeds. The narrative effect of publishing this data, rather than holding it internally, is that the early-stage ecosystem now has a shared reference point. That’s the editorial moat.
When to build an insights function — and where it sits
The advice for other companies thinking about replicating the Carta playbook is restrained. You need a density of data above some threshold — average seed-stage startup, no. You don’t need to be the market leader; you need enough flywheel to have something to talk about and a data-science function that can aggregate and anonymise. The structural question — where the function sits — Peter answers as a marketer reporting to the CMO. The unexpected note is that the insights function quickly becomes equally useful internally — sales enablement, customer success training, ecosystem-partner support — even though the work is built for outside audiences.
Energy is contagious.
The leadership tag from Peter’s former CMO Jane Alexander closes the conversation honestly. At a small startup, the founder’s energy radiates through the team because there are few enough people to feel it. As companies scale, leaders lose track of how much of the team’s engagement is just downstream of their own. The Carta insights function is, in part, an artefact of that energy applied consistently for three years — roughly how long it took to clean the data, find the questions worth asking, and become a citation rather than a publisher.
What to listen for
The full episode covers the State of Glow college-merchandise startup that taught Peter how much of entrepreneurship is sales-tax-in-26-states rather than music-festival glamour, the COVID Tracking Project chapter, the founding hypothesis behind Carta Insights, the natural-language-to-SQL caveat, the goal of making Carta data ubiquitous as the success metric, and the advice to fundraise earlier and stretch every dollar longer than 2021-instinct would suggest. His three-word descriptor is Energetic. Curious. Nerdy. Listen at /podcast/ep-023-peter-walker; for the longer thread, see /topics/talent-equity or /topics/capital-allocation.
Related questions
- What is the difference between insights and data storytelling, in Peter Walker's framing?
- Insights are the pieces of information that have utility — the aggregated result of a dataset that tells you to take an action or not take one. Data storytelling is the craft of transforming the underlying spreadsheet into charts, written headlines, and narratives that make the insight land with a specific audience. The two are sequential — there is no good data storytelling without a real insight underneath, and even a real insight wastes its power without the storytelling layer. Peter's argument is that most companies are bad at both, and the rare company is one that is good at the first and trains itself on the second.
- What does the 75% no-exercise rate on departing-employee equity at Carta tell us?
- That startup employees no longer value their equity dearly enough to pay to exercise it when they leave. About three quarters of departing employees across Carta let their incentive stock options lapse during the 90-day post-departure exercise window — meaning the only components of their compensation that became liquid were salary and bonus. The signal compounds: down valuations make exercising a bet on a security potentially worth less than the strike price; companies staying private 12 to 14 years rather than 8 to 10 means the cash outlay is locked up for longer; the option pool returns to the company and gets reissued, often in smaller equity packages to the next hires.
- Is the funding winter over, according to Carta data?
- It is thawing but still chilly, and the recovery is asymmetric across stages. Total capital raised by Carta startups peaked at about $220 billion in 2021 and was down to roughly $66 billion in 2023. Seed-stage funding is down about 30 to 35% from peak; Series D funding is down about 80%. The late-stage drought is the worrying signal because the dollar volumes are large and there are no IPOs reopening the window. Q1 2024 was slightly better than Q4 2023, but the green shoots are modest. Peter's prediction at recording: company shutdowns continue rising through the first half of 2024, as bridge and extension rounds from 2023 stop and venture capital concentrates on new investments.
- How has startup compensation reset since 2022?
- Salary compensation is approximately flat — about 0.5% higher than November 2022, which means startup salaries have not kept pace with inflation. Equity compensation has reset materially: a new hire in January 2024 received roughly 36% fewer shares than the same role hired in November 2022, and that is a literal share-count change, not a valuation effect. Engineers (especially AI/ML) and VP-and-above roles are partially insulated; the biggest impact is on early managers. The composition shift toward cash feels good in a downturn but reduces long-term upside if the company succeeds — and the option pool itself has stayed broadly stable, so the smaller equity grants reflect deliberate per-hire compression rather than pool depletion.
Updates
- Editorial pass under the v2 podcast-summary guideline.